Multilayer structured MFI-type titanosilicate: Synthesis and catalytic properties in selective epoxidation of bulky molecules

Article abstract of DOI:10.1016/j.jcat.2011.12.023

A lamellar titanosilicate (LTS-1) was hydrothermally synthesized by employing a bifunctional surfactant as the structure-directing agent (SDA). Highly crystalline LTS-1 was obtained at an optimal SDA/Si molar ratio of 0.04. As-synthesized LTS-1 possessed a multilayer structure, which was constructed from a collection of 2-nm zeolite nanosheets and interlayer SDA molecules. Removing the intercalated organic species induced irregular layer stacking to a certain extent, leading to intracrystal mesopores of ca. 3.2 nm in diameter. The catalytic performance of LTS-1 was investigated in the epoxidation of various bulky alkenes with tert-butyl hydroperoxide, cumene hydroperoxide, or aqueous H₂O₂. LTS-1 was more active than conventional titanosilicates for reactions involving bulky alkenes and oxidants, and it was immune to Ti leaching and irreversible deactivation.

Full text of DOI:10.1016/j.jcat.2011.12.023

Journal of Catalysis 288 (2012) 16–23
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Multilayer structured MFI-type titanosilicate: Synthesis and catalytic properties
in selective epoxidation of bulky molecules
⇑
Jianggan Wang, Le Xu, Kun Zhang, Honggen Peng, Haihong Wu, Jin-gang Jiang, Yueming Liu, Peng Wu
Shanghai Key Lab of Green Chemistry and Chemical Processes, Department of Chemistry, East China Normal University, North Zhongshan Rd. 3663, Shanghai 200062, People’s
Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
A lamellar titanosilicate (LTS-1) was hydrothermally synthesized by employing a bifunctional surfactant
as the structure-directing agent (SDA). Highly crystalline LTS-1 was obtained at an optimal SDA/Si molar
ratio of 0.04. As-synthesized LTS-1 possessed a multilayer structure, which was constructed from a col-
lection of 2-nm zeolite nanosheets and interlayer SDA molecules. Removing the intercalated organic spe-
cies induced irregular layer stacking to a certain extent, leading to intracrystal mesopores of ca. 3.2 nm in
diameter. The catalytic performance of LTS-1 was investigated in the epoxidation of various bulky alkenes
with tert-butyl hydroperoxide, cumene hydroperoxide, or aqueous H2O2. LTS-1 was more active than con-
ventional titanosilicates for reactions involving bulky alkenes and oxidants, and it was immune to Ti
leaching and irreversible deactivation.
Received 2 November 2011
Revised 22 December 2011
Accepted 22 December 2011
Available online 15 February 2012
Keywords:
Layered titanosilicate
TS-1
Ti-MCM-41
Ó 2012 Elsevier Inc. All rights reserved.
Mesoporous material
Epoxidation of alkenes
tert-Butyl hydroperoxide
1. Introduction
example, VPI-5 [12], UTD-1 [13], ECR-34 [14], and the ITQ series
[15–17]. However, some of these materials have poor thermal
The first titanosilicate, TS-1, containing isolated Ti ions tetrahe-
drally coordinated in the framework was discovered by Enichem in
1983 [1]. TS-1 has attracted great research interest in the past dec-
ades because of its important and unique catalytic properties for a
wide range of liquid-phase selective oxidations using hydrogen
peroxide as the oxidant, such as hydroxylation of phenol and ben-
zene, epoxidation of propylene, ammoximation of cyclohexanone
[2–5]. However, the 10-membered ring (MR) pore opening of the
MFI structure limits the applications of TS-1 to molecules with rel-
atively small dimensions. Specifically, these micropores are not
accessible to bulky substrates and organic hydroperoxides and
are easily blocked by heavy by-products. TS-1 is thus not active
for selective oxidations, which require an open reaction space.
To overcome the diffusion problems of micropores, it is desir-
able to synthesize large-pore zeolites. For this reason, 12-MR tita-
nosilicates, for example, Ti-ZSM-12 [6], Ti-Beta [7–9], Ti-MOR [10],
and Ti-MCM-68 [11], have been prepared. They were shown to be
more active than TS-1 in the epoxidation of cyclic alkenes and the
hydroxylation of aromatics with H2O2. Nevertheless, they are still
poor in oxidation using organic hydroperoxides as oxidants. The
advance of hydrothermal synthesis techniques makes available
other large-pore zeolites with 12-MR, 14-MR, or 18-MR pores, for
and hydrothermal stability, making them unsuitable for catalytic
use in liquid-phase reactions using aqueous H2O2. Moreover, it is
still a challenge to develop proper ways of incorporating transition
metal ions with redox catalytic properties into these large-pore
zeolites. On the other hand, the Ti-containing mesoporous materi-
als, for example, Ti-MCM-41 [18] and Ti-SBA-15 [19], similarly face
problems of low hydrothermal stability and low intrinsic oxidation
activity, owing to the noncrystalline and hydrophilic natures of the
framework walls. These drawbacks limit their practical applica-
tions, although Ti-MCM-41 and Ti-SBA-15 are considerably attrac-
tive for processing large molecules.
Molecular transport and diffusion in zeolite channels can be
improved by reducing zeolite particle size or creating intracrystal
mesopores. The chemical treatments of dealumination and desili-
cation [20–22] and hard and soft templating techniques [23–26]
have been employed to create defect sites and mesopores. This
improves the accessibility of active sites, and thus enhances the
catalytic activity to a certain extent. These structural modifications
take place mostly on a submicrometer scale rather than at a unit
cell level, which may prevent the catalytic ability of zeolites from
being fully utilized.
Recently, the population of zeolites that derive from lamellar
precursors has been increasing [27]. These so-called layered zeo-
lites are possibly subject to postmodifications and thus exhibit
structural diversities. Facile techniques have been developed to
⇑
Corresponding author. Fax: +86 21 6223 2292.
E-mail address: pwu@chem.ecnu.edu.cn (P. Wu).
0021-9517/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2011.12.023

J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
17
modify the structure of layered MCM-22 aluminosilicate and Ti-
MWW titanosilicate. The methods include intercalation or pillaring
with amorphous silica [27], full or partial phase delamination by
exfoliating the zeolite layers [28–30], and interlayer silylation with
organosilane [31]. These modifications enlarge the interlayer pore
entrance effectively and improve the accessibility of the active
sites located inside the internal channels. These structural changes
take place at a unit cell level, leading to catalytic materials applica-
ble to large substrates. With respect to MFI-type TS-1, the synthe-
sis of a lamellar precursor is thus expected.
Recently, Choi et al. developed a facile and controllable route for
direct synthesis of ultrathin ZSM-5 crystals that are composed of
nanosheets with 2-nm thickness by using a bifunctional surfactant
as SDA [32]. The lamellar ZSM-5 aluminosilicates, either with a
multilayer or a single-layer structure, exhibit much better catalytic
activity and selectivity than conventional bulk crystals in acid-
catalyzed reactions involving bulky molecules. In addition, they
are highly resistant to deactivation and possess excellent thermal
and hydrothermal stability. Very recently, the synthesis has been
extended to the preparation of MFI titanosilicate nanosheets with
a thickness of a single-unit cell [33], which were shown to be
highly efficient in the oxidations involving bulky reactants.
Herein, following this pioneering research, we have carried out
the synthesis of lamellar TS-1 (LTS-1) with a multilayer structure
simply by adding a bifunctional surfactant into the synthetic sys-
tem of conventional TS-1. We have investigated the effect of
bifunctional surfactant content on the formation of LTS-1 and
extensively evaluated its catalytic properties in the epoxidation
of various alkenes with tert-butyl hydroperoxide, cumene hydro-
peroxide, or aqueous H2O2 as the oxidant.
Ti-MCM-41 with a hexagonal mesostructure was synthesized at a
Si/Ti ratio of 40 according to a literature method [18].
2.3. Characterization methods
The X-ray powder diffraction (XRD) patterns were measured on
a Rigaku Ultima IV X-ray diffractometer using Cu K
a radiation
(k = 1.5405 Å). Scanning electron microscopy (SEM) was performed
on a Hitachi S-4800 microscope, while the TEM images were taken
on a JEOL-JEM-2100 microscope. The Ti content actually incorpo-
rated into the final titanosilicates was determined by inductively
coupled plasma emission spectrometry (ICP) on a Thermo IRIS In-
trepid II XSP atomic emission spectrometer after the sample was
dissolved in aqueous HF solution. The specific surface area and pore
size distribution were measured by N2 adsorption at 77 K on a BEL-
SORPMAX instrument equipped with low-pressure sensors. Before
adsorption, the samples were activated under vacuum at 573 K for
10 h. The pore size distribution of micropores was calculated by the
Horvath–Kawazoe (HK) method, whereas that of mesopores was
calculated by the BJH method. The UV–visible diffuse reflectance
spectra were recorded on a Perkin–Elmer UV/vis Lambda25 spec-
trometer using BaSO4 as a reference. The IR spectra were collected
on a Nicolet Nexus 670 FT-IR spectrometer in absorbance mode at a
spectral resolution of 2 cmÀ1. The spectra in the region of zeolite
framework vibration were measured using the KBr technique. The
spectra of hydroxyl stretching were measured on self-supported
wafers (2 cm in diameter and 4.8 mg cmÀ2 in thickness). The sam-
ples were evacuated for 2 h at 723 K in a quartz IR cell sealed with
CaF2 windows, which were connected to a vacuum system. All spec-
tra were collected at room temperature. 29Si solid-state MAS NMR
spectra were recorded on a VARIAN VNMRS-400WB spectrometer
under a one-pulse condition. The spectra were obtained with a fre-
quency of 79.43 MHz, a spinning rate of 3.0 kHz, and a recycling
delay of 60 s. The chemical shift was referred to Q8M8 ([(CH3)3SiO]8-
SiO12). Thermogravimetric analysis (TGA) was measured on a MET-
TLER TOLEDO TGA/SDTA851e in air.
2. Experimental
2.1. Synthesis of bifunctional surfactant
The
linear
bifunctional
structure-directing
agent
(C22H45AN(CH3)2AC6H12AN(CH3)2AC6H13)Br2 (denoted as C22-6-
6Br2) was synthesized following the procedures described in the
literature [32]. The as-synthesized diammonium salt C22-6-6Br2
was converted into hydroxide form, C22-6-6(OH)2, through reaction
with Ag2O in distilled water. The solution containing C22-6-6Br2,
Ag2O, and water was stirred for 20 h at an Ag2O/C22-6-6Br2 molar
ratio of 1.2 at room temperature. The remaining Ag2O powder
and the AgBr precipitate formed were removed by filtration.
2.4. Epoxidation reactions
The epoxidation of alkenes with H2O2 (30 wt.% aqueous solu-
tion) or anhydrous tert-butyl hydroperoxide (TBHP, 5.5 M in dec-
ane) was carried out in a 25-mL round-bottomed flask equipped
with reflux condenser. The alkenes included 1-hexene, cyclohex-
ene, cyclooctene, and cyclododecene. In a typical run, 50 mg of cat-
alyst, 10 mL of acetonitrile, 10 mmol of alkene, and 10 mmol of
H2O2 or TBHP were mixed in the flask and stirred vigorously at
333 K for 2 h. The epoxidation of propylene with cumene hydro-
peroxide (CMHP) was carried out in a 50-mL Teflon-lined autoclave
reactor. First, 50 mg of catalyst, 1 mmol CMHP, and 10 mL of cu-
mene were added to the reactor, and it was then charged with ex-
cess propylene at 0.5 MPa. The reaction was carried out under
vigorous stirring at 373 K for 2 h. All products were analyzed on
a Shimadzu GC-2014 gas chromatograph equipped with a DM-
Wax-30m column (Dikma Technologies Inc.) and an FID detector.
2.2. Synthesis of Ti-containing materials
In a typical synthesis of lamellar TS-1, tetraethyl orthosilicate
(TEOS) and tetrabutyl orthotitanate (TBOT) were hydrolyzed in an
aqueous solution of C22-6-6(OH)2 without removal of ethanol, form-
ing a synthetic gel with a molar composition of 1.0 SiO2:0.0167
TiO2:0.02–0.16 C22-6-6(OH)2:100 H2O:4.0 EtOH. The resultant gel
was transferred to a Teflon-lined stainless steel autoclave, which
was heated at 423 K for 20 days under rotation (60 rpm). After crys-
tallization, the solid product was filtered, washed with distilled
water, and dried at 393 K. The product was further calcined to burn
off the occluded organic species at 823 K for 10 h in air.
For control experiments, conventional TS-1 (CTS-1) was synthe-
sized following the classical method developed by ENI [1]. The
hydrothermal synthesis was carried out using tetrapropylammoni-
um hydroxide (TPAOH) as SDA at a TPAOH/SiO2 ratio of 0.18 and a
Si/Ti ratio of 40 at 443 K for 48 h. Ti-MWW (Si/Ti = 38) [34] and
Al-free Ti-Beta (Si/Ti = 40) [7] were hydrothermally synthesized
following the procedures reported previously. Mesoporous
3. Results and discussion
3.1. Synthesis and characterizations of multilayer structured
titanosilicate
Differently from TPAOH, the conventional SDA for crystallizing
the MFI zeolites, the bifunctional surfactant present in the gels
greatly influenced the crystallization processes for the MFI phase.
The effect of the amount of bifunctional SDA on the crystallization,
as well as the characteristics of the products, was thus investigated

18
J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
by varying the C22-6-6(OH)2/Si ratio while fixing the Si/Ti ratio at 60
in the synthetic gel. The crystallinity of as-synthesized samples
was characterized by XRD (Fig. 1). XRD patterns revealed that
the samples showed the characteristic MFI diffraction peaks at
7.95°, 8.94°, 23.2°, and 24.1°, which are assigned to the [101],
[200], [501], and [303] planes, respectively. At 0.04 6 C22-6-6
(OH)2/Si 6 0.08 (Fig. 1c and d), the absence of any baseline drift
and amorphous phase-related broad diffraction around 21.5° im-
plied that the samples were highly crystalline materials. The dif-
fraction peaks due to the crystalline TiO2 phase were not
observed in XRD patterns, suggesting that the Ti species did not
condense to form oxides, but most possibly were highly dispersed
and incorporated into the MFI framework. However, at lower or
higher C22-6-6(OH)2/Si ratios, the crystallization to the MFI phase
tended to be incomplete, resulting in the products with amorphous
phase coexisting (Fig. 1b, e, and f). Thus, to obtain pure LTS-1 with
high crystallinity, the C22-6-6(OH)2/Si ratio in the synthetic gel
should be controlled between 0.04 and 0.08. In comparison to con-
ventional TS-1 synthesized with TPAOH (Fig. 1a), bifunctional sur-
factant-assisted LTS-1 materials were less intensive in diffraction
peaks, and exhibited the behavior of oriented growth in the crys-
tals. The [0k0] diffractions were not observed in the XRD patterns
of the LTS-1 samples, which indicated that their crystalline
structure lacked long-range order along the b axis. Moreover, the
LTS-1 samples were characteristic of a mesostructure in showing
a typical diffraction at 1.25–1.73°. This diffraction, however, disap-
peared in the low-angle region after the occluded organic species
were burned off in air at 823 K for 10 h (not shown). This means
that the mesophase was of a soft or changeable structure.
The crystallinity and morphology of LTS-1 synthesized at a C22-
6-6(OH)2/Si ratio of 0.04 in comparison with conventional TS-1
were investigated by scanning electron microscopy. The crystals
of CTS-1 showed the typical ‘‘blackberry’’ morphology and had a
quite uniform crystal size of 300–400 nm (Fig. 2a). As illustrated
in the inset figure, these crystals were secondary particles that con-
sisted of much smaller irregular nanocrystals. This kind of crystal
morphology is essential to guarantee high catalytic activity of TS-
1, as the interruption between nanocrystals would shorten the dif-
fusion path for both reactants and products penetrating through
zeolites. Nevertheless, this kind of nanocrystal assembly provides
only a limited number of intracrystal voids or large pores, which
are still not enough to produce high catalytic activity for bulky
molecules.
(OH)2 replacing TPAOH was consistent with the XRD investigation,
suggesting that LTS-1 possessed a lamellar structure. As demon-
strated clearly in the literature [32], these phenomena are closely
related to the bifunctionality of C22-6-6(OH)2. The hydrophilic
groups of quaternary diammonium cations would interact readily
with the negatively charged surface of the silicate formed in basic
media and then serve as the SDA for crystallizing the MFI structure,
whereas the long chain of alkyl groups with a hydrophobic nature
may act as a spacer to prevent the zeolite nanocrystals or nano-
sheets from growing continuously. Moreover, the interaction be-
tween the long-chain tails induced the growth of crystals having
the ac plane orientation, leading to a lamellar material.
High-resolution transmission electron microscope (TEM)
images provided more reliable information about the structure of
LTS-1 at a unit cell scale. The images of as-synthesized LTS-1 re-
vealed that it was composed of stacks of 2-nm alternating layers
and 2.7-nm surfactant micelles (Figs. 3a and b). The layer thickness
equals to the unit cell parameter of the MFI zeolite along the b axis.
In TG-DTG analysis, LTS-1 showed weight loss in the temperature
window of 423–873 K, revealing that the surfactant content was
about 42 wt.% in as-synthesized product. This was in agreement
with the results previously reported on multilamellar ZSM-5 alumi-
nosilicates [32]. Thus, as-synthesized LTS-1 is considered to have a
multilayer structure, as graphically illustrated in Scheme 1. Never-
theless, it was recently reported that the titanosilicate with a sin-
gle-unit cell structure was crystallized from the synthetic system
employing C16-6-6(OH)2 with a shorter alkyl group as the SDA and
after removing the ethanol formed from TEOS hydrolysis [33].
Calcination in air at 823 K completely removed the occluded sur-
factant. XRD patterns indicated that the ordered mesostructure
owing to layer stacking was destroyed. This implied that dehydr-
oxylation occurred, with the intercalated organic species leaving
of the interlayer spaces, which in turn created SiAOASi linkages
mending the interrupted zeolite structure. However, the long-range
ordered structure along the direction of layer stacking, that is, the b
axis, was not restored perfectly. The mismatching of layer conden-
sation induced intracrystal faults and defects, leading to the forma-
tion of mesopores. As indicated by the arrows in Fig. 3c and d, the
mesopores were present within many plate crystals. Simulta-
neously, Fig. 3d also shows the image of the [010] plane in the
upper left corner, representing the pore windows of straight
10-MR channels. Obviously, the pore array was regular and ordered
without interruption on the [010] plane. Thus, the mesopores ex-
isted in LTS-1 predominantly between the unit cells along the b
direction.
N2 adsorption was then carried out to investigate the textural
properties of LTS-1. Fig. 4 shows its adsorption/desorption iso-
therms and pore size distributions, together with those of CTS-1
and Ti-MCM-41. The isotherms of LTS-1 combined the character-
istics of type I and type IV (Figs. 4A, b). It had an adsorption
amount comparable to that for CTS-1 at P/P0 < 0.1, which was
mainly attributed to the monolayer adsorption within 10-MR
micropores. However, LTS-1 showed a much higher N2 adsorption
amount than CTS-1 in the region of higher relative pressure. The
hysteresis loop at P/P0 > 0.5 could be assigned to the type H4,
implying the presence of mesopores formed by the aggregation
of nanosheets. This was further verified by the results given by
pore size distributions (Fig. 4B, b). LTS-1 was a hybrid material
that possessed both micropores of ca. 0.5 nm and mesopores with
a narrow distribution centered at 3.2 nm. The former is attributed
to the 10-MR channels of the MFI structure, whereas the latter
should be the contribution of the intracrystal mesopores owing
to the irregular layer stacking, as evidenced by TEM images. On
the other hand, CTS-1 and Ti-MCM-41 showed a single pore distri-
bution, possessing only micropores and mesopores, respectively
(Fig. 4B, a and c).
On the other hand, the SEM image indicated that LTS-1 was also
a crystalline material containing no amorphous phase (Fig. 2b).
Quite differently from CTS-1, the LTS-1 crystals exhibited a thin
plate-like morphology. The crystals were about 20–50 nm in thick-
ness and 0.5–1
lm in length. The morphology change with C22-6-6
Amorphous phase
f
e
d
c
b
*1/3
a
1.0 2.0 3.0
5
10
15
20
25
30
2Theta (degree)
Fig. 1. XRD patterns of CTS-1 (a) and LTS-1 samples synthesized at C22-6-6(OH)2/Si
ratios of 0.02 (b), 0.04 (c), 0.08 (d), 0.12 (e), and 0.16 (f).

J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
19
a
b
0.2 µm
0.5 µm
2 µm
1 µm
Fig. 2. SEM images of CTS-1 (a) and LTS-1 synthesized at C22-6-6(OH)2/Si = 0.04 after calcination (b). The insets show enlarged images.
c
d
a
b
b
a
Surfactant micelle
2.7 nm
MFI sheet
2 nm
b
a
b
a
Fig. 3. TEM images of LTS-1 as-synthesized at C22-6-6(OH)2/Si = 0.04 (a and b) and after calcination (c and d). The arrows indicate the presence of mesopores.
The coordination states of Ti species are determinative of the
catalytic activity of titanosilicates. FTIR spectroscopy was first
employed to investigate the Ti states. As shown in Fig. 5, Ti-free
Silicalite-1, CTS-1, and LTS-1 all showed an intensive band at
554 cmÀ1, which is assigned to the vibration of double five-mem-
bered ring units and is considered as the fingerprint of the MFI
structure [35]. Both CTS-1 and LTS-1 showed a characteristic band
at 967 cmÀ1 that is attributed to a collective vibration of the
SiAOATi bond or SiAO bond perturbed by the presence of Ti atoms
in the framework of TS-1 [35]. Meanwhile, Silicalite-1 only showed
an extremely weak band at 956 cmÀ1, which can probably be as-
signed to the silanol groups related to the defect sites [34,36]. The
big difference from Silicalite-1 and the similarity to TS-1 strongly
supported the conclusion that LTS-1 contained the isolated Ti(IV)
species with a tetrahedral coordination in the skeleton of the MFI
zeolite.
Scheme 1. Multilayered MFI titanosilicate directed by bifunctional SDA.
The coordination states of Ti species in LTS-1 were further stud-
ied by UV–visible spectroscopy (Fig. 6). Similar to CTS-1 and
Ti-MCM-41, LTS-1 showed a main absorption around 210 nm, which
is assigned to the charge transfer from O2À to Ti4+. This adsorption is
usually observed for Ti-substituted zeolites and is characteristic of
tetrahedrally coordinated Ti highly dispersed in the framework
[35]. In agreement with these IR spectra, UV–vis spectra also verified
that LTS-1 contained isolated Ti species. On the other hand, besides
The textural properties of the samples are summarized in
Table 1. LTS-1 had a higher BET surface area and a larger mesopore
volume than CTS-1. Moreover, the proportion of external surface
area to overall surface area was 70% for LTS-1, indicating that the
mesopores in LTS-1 made a great contribution to the specific surface
area and total pore volume. The mesoporosity thus created is ex-
pected to alleviate the diffusion limitation effectively when the
materials are applied to catalysis and separation.
the absorption around 210 nm, Ti-MCM-41 showed
a broad

20
J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
600
500
400
300
200
100
0
A
B
c
b
c
b
a
a
0.5
1.0
4
8
12
0.0
0.2
0.4
0.6
0.8
1.0
Pore diameter (nm)
P/P0
Fig. 4. N2 adsorption isotherms at 77 K (A) and pore size distribution (B) of CTS-1 (a), LTS-1 (b), and Ti-MCM41 (c). The isotherm is offset vertically by 100 cm3 gÀ1 for Ti-
MCM-41.
Table 1
Textural properties of various titanosilicates.
6
b
c
Sample
Pore size PVmicropa(cm3 gÀ1
(nm)
)
PVmesop
SBET
Sext
210
(cm3 gÀ1
)
(m2 gÀ1
)
(m2 gÀ1
)
5
4
3
2
1
0
260
LTS-1
CTS-1
Ti-MCM-41 2.9
0.5, 3.2
0.5
0.15
0.16
0.06
0.45
0.08
0.50
510
425
890
357
73
860
a
Micropore volume was calculated by t-plot method.
Mesopore volume = total pore volume À micropore volume. Total pore volume
b
was obtained at P/P0 of 0.99.
c
c
External surface area was calculated by t-plot method.
b
a
1.5
1.0
200
300
400
500
Wavelength (nm)
Fig. 6. UV–visible spectra of CTS-1 (a), LTS-1 (b), and Ti-MCM-41 (c).
lytic performance of titanosilicates, as the reactions are usually
conducted in the presence of water when aqueous H2O2 is em-
ployed as the oxidant. IR spectra in the region of hydroxyl stretch-
ing were thus measured to investigate this issue. As shown in
Fig. 7, both CTS-1 and LTS-1 showed bands at 3745, 3690, and
3500 cmÀ1. They are attributed to the terminal silanol groups,
geminal Si(OH)2 groups, and hydrogen-bonded silanol groups on
hydroxyl nests [37,38]. The 3745 cmÀ1 band of LTS-1 was more
intensive and sharp than that of CTS-1. This is presumed to be
associated with a larger surface on the mesopores in LTS-1. Thus,
LTS-1 may have a more hydrophilic surface. On the other hand,
the 3690 and 3500 cmÀ1 bands related to the defect sites were
comparable between CTS-1 and LTS-1.
0.5
c
b
a
0.0
1000 900
800
700
600
500
Wavenumber (cm-1
)
Fig. 5. FTIR spectra in the region of framework vibration for Ti-free Silicalite-1 (a),
CTS-1 (b), and LTS-1 (c).
The structural configuration and the states of Si species were
further investigated by 29Si MAS NMR. As depicted in Fig. 8, the
29Si NMR spectra of CTS-1 and LTS-1 both showed a main reso-
nance at À113 ppm and a weak resonance peak at À103 ppm,
which are attributed to Si(OSi)4 (Q4) and Si(OSi)3OH (Q3) species,
respectively [39]. Both samples almost lacked the Si(OSi)2OH2
(Q2) species at À92 ppm. The Q4/Q3 ratio calculated from the
deconvoluted peaks was 7.5 and 6.3 for CTS-1 and LTS-1, respec-
tively. The higher percentage of Q3 verified again that LTS-1 con-
adsorption at 230–270 nm, implying the coexistence of part of the
penta- or hexacoordinated Ti species as a result of water adsorption
[34]. This is probably related to the hydrophilic nature of the amor-
phous framework walls in Ti-MCM-41. Although similarly contain-
ing mesopores, LTS-1 was basically different from the pure
mesoporous material Ti-MCM-41 and was closer to crystalline zeo-
lites in nature.
In addition to crystal size, porosity, and Ti states, the hydropho-
bic/hydrophilic properties also play an important role in the cata-
tained more hydroxyl groups and thus
surface than CTS-1.
a more hydrophilic

J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
21
Table 2
1.0
0.8
0.6
0.4
0.2
0.0
Epoxidation of cyclohexene with TBHP over various titanosilicates.a
3745
d
No. Catalyst
Si/Tib Conv. (%) CHO sel. (%)c TOF (hÀ1
)
Eff. (%)e
1
2
3
4
5
CTS-1
41
40
38
58
0.1
5.3
4.2
16.7
18.1
95.0
99.0
99.2
99.2
99.3
0.2
13.0
9.8
61.9
45.3
93.2
89.5
90.8
92.7
86.1
Ti-Beta
Ti-MWW
LTS-1
Ti-MCM-41 41
a
Reaction conditions: catalyst, 50 mg; MeCN, 10 mL; cyclohexene, 10 mmol;
TBHP (5.5 M in decane), 10 mmol; temp., 333 K; time, 2 h.
b
a
b
Molar ratio determined by ICP.
Cyclohexene oxide (CHO).
Turnover frequency (TOF): amount of cyclohexene converted per Ti site per
c
d
hour.
e
Efficiency of TBHP = (amount of product/amount of TBHP consumed) Â 100.
3800
3600
3400
3200
3000
Wavenumber (cm-1
)
and Ti-MCM-41 was always over 99% independent of reaction time
(Fig. S1). It is deduced that the active Ti sites in LTS-1 were highly
accessible to the bulky molecules because of the presence of
mesopores.
Fig. 7. FTIR spectra in the hydroxyl stretching region of CTS-1 (a) and LTS-1 (b).
The epoxidation of cyclohexene with aqueous H2O2 is useful for
investigating the usability of the Ti active sites located on the sur-
faces and pore entrances of zeolite crystals. The reaction involves
three pathways, isolated Ti species-catalyzed catalytic epoxidation
of C@C bonds to CHO, noncatalytic or radical oxidation on the allylic
positions to 2-cyclohexene-1-ol and 2-cyclohexene-1-one, and
hydration of CHO with water over the weak acid sites in titanosili-
cate to the secondary by-product 1,2-cyclohexanediol. Table 3 com-
pares the results observed on LTS-1, CTS-1, and Ti-MCM-41. In
comparison to epoxidation with TBHP, using aqueous H2O2 greatly
retarded the reaction. It seemed that a hydrophilic microenviron-
ment where the Ti species were situated was not favorable for reac-
tions in the presence of water. Similar phenomena have been
reported on CTS-1 catalysts with mesoporous/microporous hierar-
chical structures obtained in the presence of amphiphilic organosi-
lanes [41,42]. A highly hydrophilic character may inhibit molecular
diffusion inside the pores, especially in the presence of water. Nev-
ertheless, taking advantage of mesoporosity, LTS-1 still displayed a
higher conversion and TOF for cyclohexene than TS-1. And it was
more selective to CHO than Ti-MCM-41 with a high hydrophilicity.
4
Q
3
Q
b
a
2
Q
-90
-100
-110
-120
Chemical shift (ppm)
Fig. 8. 29Si MAS NMR spectra of CTS-1 (a) and LTS-1 (b).
3.2. Catalytic properties of LTS-1 in alkene epoxidations
3.2.2. Epoxidation of 1-hexene with H2O2
The epoxidation of 1-hexene with H2O2 requires relatively nar-
row reaction spaces. Corresponding activity would be useful to
evaluate the catalytic properties of the Ti sites located within the
micropores. Hydrophilic Ti-MCM-41 due to abundant silanol
groups in amorphous silica walls was almost inactive to this reac-
tion, independent of the solvent used (Table 4, No. 3). Although
LTS-1 was less active than CTS-1, the conversion achieved on
LTS-1 indicated that its catalytic behavior was still that character-
istic of crystalline titanosilicates rather than Ti-containing amor-
phous materials. LTS-1 showed the same solvent effect as CTS-1;
that is, the conversion of 1-hexene was higher in MeOH than in
MeCN. This was in agreement with the literature [43], indicating
that LTS-1 retained the catalytic features of CTS-1. The lower activ-
ity of LTS-1 in H2O2 may be because it contained a larger number of
silanol groups than CTS-1, as LTS-1 showed a more intensive band
at 3744 cmÀ1 in IR spectrum (Fig. 7) and a smaller Q4/Q3 ratio in
29Si NMR spectrum (Fig. 8). In the presence of aqueous hydrogen
peroxide, water molecules would adsorb onto the surface and
internal silanols. This makes the Ti sites not easily accessible to
the substrates and thus reduces the catalytic activity.
3.2.1. Epoxidation of cyclohexene with TBHP and H2O2
The epoxidation activity of LTS-1 has been investigated by com-
paring it with that of crystalline TS-1, Ti-MWW, and Ti-Beta, with
only micropores, as well as that of Ti-MCM-41, with ordered mes-
opores but amorphous walls. Cyclic alkenes, TBHP, and CMHP were
chosen as the substrates or oxidants in order to evaluate the Ti
sites largely exposed on the external surface because these mole-
cules are too large to diffuse into the 10-MR pores of the MFI struc-
ture [40].
The epoxidation of cyclic alkenes was carried out in acetonitrile,
as it was shown to be a more suitable solvent than methanol for
TS-1 [41]. In the epoxidation of cyclohexene with TBHP (5.5 M in
decane), cyclohexene oxide (CHO) was almost the sole product,
irrespective of titanosilicate types and reaction time (Table 2 and
Fig. S1). CTS-1 showed the lowest conversion of cyclohexene be-
cause the 10-MR pores of the MFI structure provided a significant
hindrance to the bulky molecules of cyclohexene and TBHP (Table
2, No. 1). Even when the reaction time was prolonged to 80 h, the
conversion was still very low (Fig. S1). Having larger pores of 12-
MR, Ti-Beta and Ti-MWW showed a higher conversion and a higher
specific activity (TOF) than CTS-1 (Table 2, Nos. 2 and 3). Notably,
the conversion of cyclohexene on LTS-1 was much higher than on
CTS-1, Ti-Beta, and Ti-MWW. Its TOF reached the same level as on
Ti-MCM-41 (Table 2, Nos. 4 and 5). The epoxide selectivity of LTS-1
3.2.3. Epoxidation of various cyclic alkenes with TBHP
These investigations verified that LTS-1 possessed a chief asset
that it enabled the epoxidation of bulky alkenes by taking

22
J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
Table 3
Epoxidation of cyclohexene with H2O2 over various titanosilicates.a.
b
c
No.
Catalyst
Si/Ti
Conv. (%)
TOF (hÀ1
)
Eff.d (%)
Product sel.e (%)
CHO
Allylic
Diol
1
2
3
LTS-1
CTS-1
Ti-MCM-41
58
41
41
4.0
2.8
5.2
15.3
7.2
13.1
85.6
87.8
78.7
19.4
22.4
5.6
76.4
75.6
93.4
4.2
2.0
1.0
a
b
c
Reaction conditions: catalyst, 50 mg; MeCN, 10 mL; cyclohexene, 10 mmol; H2O2 (30 wt.% aqueous solution), 10 mmol; temp., 333 K; time, 2 h.
Molar ratio determined by ICP.
Turnover frequency.
d
e
Efficiency of H2O2 = (amount of product/amount of H2O2 consumed) Â 100.
CHO, cyclohexene oxide; allylic oxidation products, 2-cyclohexene-1-ol and 2-cyclohexene-1-one; diol, 1,2-cyclohexanediol.
Table 4
Epoxidation of 1-hexene with H2O2 over various titanosilicates.a
Table 6
Reuse of LTS-1 and Ti-MCM-41 in epoxidation of cyclohexene with H2O2.a.
No.
Catalyst
Si/Tib
CH3CN (%)
MeOH (%)
Catalyst
LTS-1
Reuse number
Conv. (%)
Product sel.b (%)
d
c
d
Conv.
Sel.c
Eff.
Conv.
Sel.
Eff.
CHO
Allylic
Diol
1
2
3
LTS-1
CTS-1
Ti-MCM-41
58
41
41
4.1
12.1
0.2
99.6
99.4
99.1
74.4
75.7
70.9
6.5
25.2
0.4
95.2
92.1
90.3
82.3
84.5
73.1
1
2
3
4
5
17.2
17.0
17.2
16.9
17.3
36.5
37.4
37.5
36.8
37.0
44.5
43.1
43.3
44.2
43.7
19.0
19.5
19.2
19.0
19.3
a
Reaction conditions: catalyst, 50 mg; MeCN or MeOH, 10 mL; 1-hexene,
10 mmol; H2O2 (30 wt.% aqueous solution), 10 mmol; temp., 333 K; time, 2 h.
b
Molar ratio determined by ICP.
The main by-product is diol.
Efficiency of H2O2 = (amount of product/amount of H2O2 consumed) Â 100.
Ti-MCM-41
1
2
3
4
5
18.5
14.3
12.7
10.2
7.6
21.3
15.3
10.1
10.7
8.5
59.3
63.5
73.5
74.6
76.6
19.4
21.2
16.4
14.7
14.9
c
d
advantage of the mesoporosity deriving from the multilayer pre-
cursor. To show that this effect was not limited to cyclohexene,
LTS-1 was tested for the epoxidation of other bulky alkenes with
larger molecular dimensions (Table 5). LTS-1 was highly active
with all the substrates investigated, indicating that the Ti sites
were also accessible to reactants larger than cyclohexene. This
should be the advantage of the special pore architecture and crys-
tal morphology in LTS-1. Interestingly, LTS-1 even showed a higher
conversion for cycloheptene and cyclooctene than for cyclohexene
(Table 5, Nos. 1–3). As long as the Ti sites can be reached by the
substrates, the catalytic activity will be closely related to the reac-
tivity of the substrates themselves. As the C@C bonds in cyclohep-
tene and cyclooctene have a higher electrophilicity than those in
cyclohexene, and their corresponding epoxides are structurally
more stable in comparison with the more strained cyclohexene
epoxide [44], it is reasonable that LTS-1 was more active to the
former.
a
Reaction conditions: catalyst, 100 mg; cyclohexene, 5 mmol; H2O2 (30 wt.%
aqueous solution), 2.5 mmol; MeCN, 5 mL; temp., 333 K; time, 2 h. The spent cat-
alyst was regenerated by calcination in air at 773 K for 5 h after each run.
b
CHO, cyclohexene oxide; allylic oxidation products, 2-cyclohexene-1-ol and 2-
cyclohexene-1-one; diol, 1,2-cyclohexanediol.
MCM-41. The spent catalyst was calcined in air at 773 K for 5 h,
burning off any organic species occluded. The activated catalyst
was then subjected to repeated epoxidation. In the case of LTS-1,
the regeneration between two reaction runs nearly restored the
initial activity. We also checked the catalytic activity of the filtrate
obtained after refluxing LTS-1 in CH3CN. The filtrate was essen-
tially inactive for the epoxidation of cyclohexene, indicating no
leaching of Ti into the reaction medium. LTS-1 was thus a stable
and robust catalyst. On the other hand, the conversion and epoxide
selectivity on Ti-MCM-41 both decreased gradually in repeated
reactions irrespective of the calcination regeneration (Table 6).
Since the calcination readily removed the heavy by-products, pos-
sibly depositing them inside pores or covering the Ti active sites,
the decrease in conversion implied that Ti-MCM-41 suffered irre-
versible deactivation. This is presumed to be associated with
unstable Ti species existing in amorphous silica walls [19,45]. In
fact, UV–visible investigation confirmed that the Ti coordination
state in used LTS-1 catalyst was essentially the same as that in
fresh catalyst (Fig. S2A). On the other hand, after the fourth reuse,
Ti-MCM-41 showed a UV–visible spectrum with a main band
around 260 nm (Fig. S2B), indicating that the Ti species in Ti-
MCM-41 changed irreversibly from tetrahedral coordination to
octahedral coordination after catalytic reuse.
3.2.4. Reusability of LTS-1
In terms of processing bulky molecules, LTS-1 was characteris-
tic of mesoporous materials (Tables 2 and 5). We found that its sta-
bility and reusability were very close to those of crystalline
microporous titanosilicates. Table 6 compares catalytic reuse in
the epoxidation of cyclohexene with H2O2 between LTS-1 and Ti-
Table 5
Epoxidation of cyclic alkenes with TBHP over LTS-1 catalyst.a
No. Substrate
Conv. (%) Epoxide sel. (%) TOFb (hÀ1
)
Eff.c (%)
1
2
3
4
Cyclohexene
16.7
18.6
26.5
8.5
99.2
99.3
99.7
99.4
61.9
68.9
97.0
31.1
92.4
94.6
95.8
90.9
Cycloheptene
Cyclooctene
3.2.5. Epoxidation of propylene with cumene hydroperoxide
Cyclododecened
LTS-1 showed unique activity for the epoxidation of alkenes,
which requires relatively open reaction spaces. This encouraged
us to investigate its potential application to the production of pro-
pylene oxide (PO). Sumitomo Chemical Co. Ltd. has already com-
mercialized an innovative process for PO production using
cumene hydroperoxide (CMHP) as the oxidant [46]. As shown in
a
Reaction conditions: LTS-1, 50 mg; MeCN, 10 mL; alkexene, 10 mmol; TBHP
(5.5 M in decane), 10 mmol; temp., 333 K; time, 2 h.
b
Turnover frequency (TOF), amount of cyclohexene converted per Ti site per
hour.
c
Efficiency of H2O2 = (amount of product/amount of TBHP consumed) Â 100.
d
Mixture of cis and trans isomers.

J. Wang et al. / Journal of Catalysis 288 (2012) 16–23
23
Acknowledgments
We gratefully acknowledge the NSFC of China (U1162102,
20925310, and 20873043), the Science and Technology Commis-
sion of Shanghai Municipality (09XD1401500), and the Shanghai
Leading Academic Discipline Project (B409).
Appendix A. Supplementary material
Scheme 2. Innovative CHPO process for PO production.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2011.12.023.
Table 7
Epoxidation of propylene with CMHP over various titanosilicates.a
References
[1] M. Taramasso, G. Perego, B. Notari, US Patent 4410501, 1983.
[2] G. Bellussi, M.S. Rigutto, Stud. Surf. Sci. Catal. 137 (2001) 911.
[3] P. Ratnasamy, D. Srinivas, H. Knözinger, Adv. Catal. 48 (2004) 1.
[4] L.J. Davies, P. McMorn, D. Bethell, P.C.B. Page, F. King, F.E. Hancock, G.J.
Hutchings, J. Catal. 198 (2001) 319.
No.
Catalyst
Si/Tib
Conv. (%)
PO sel. (%)
PO yield (%)
1
2
3
4
5
CTS-1
Ti-Beta
Ti-MWW
LTS-1
Ti-MCM-41
41
40
38
58
41
0.8
2.0
0.8
20.0
7.3
99.3
99.1
99.5
99.7
99.2
0.80
1.98
1.79
19.94
7.24
[5] A. Bhaumik, P. Mukherjee, R. Kumar, J. Catal. 178 (1998) 101.
[6] A. Tuel, Zeolites 15 (1995) 236.
[7] T. Blasco, M.A. Camblor, A. Corma, P. Esteve, J.M. Guil, A. Martínez, J.A.
Perdigón-Melón, S. Valencia, J. Phys. Chem. B 102 (1998) 75.
[8] J. Le Bars, J. Dakka, R.A. Sheldon, Appl. Catal. A Gen. 136 (1996) 69.
[9] T. Tatsumi, N. Jappar, J. Phys. Chem. B 102 (1998) 7126.
[10] P. Wu, T. Komatsu, T. Yashima, J. Phys. Chem. 100 (1996) 10316.
[11] Y. Kubota, Y. Koyama, T. Yamada, S. Inagaki, T. Tatsumi, Chem. Commun. 46
(2008) 6224.
a
Reaction conditions: catalyst, 50 mg; cumene, 10 mL; CMHP, 1 mmol; propyl-
ene, 0.5 MPa temp., 373 K, time, 2 h.
b
Molar ratio determined by ICP.
[12] M.E. Davis, C. Saldarriaga, C. Montes, J. Garces, C. Crowdert, Nature 331 (1988)
698.
[13] C.C. Freyhardt, M. Tsapatsis, R.F. Lobo, K.J. Balkus, M.E. Davis, Nature 381
(1996) 295.
[14] K.G. Strohmaier, D.E.W. Vaughan, J. Am. Chem. Soc. 125 (2003) 16035.
[15] A. Corma, M.J. Diaz-Cabanas, F. Rey, S. Nicolopoulus, K. Boulahya, Chem.
Commun. (2004) 1356.
[16] A. Corma, M.J. Diaz-Cabanas, J. Martinez-Triguero, F. Rey, J. Rius, Nature 418
(2002) 514.
Scheme 2, this so-called CHPO (cumene hydroperoxide propylene
oxide) technique avoids co-producing other organic substances,
since cumene serving as an oxygen carrier is recycled via two redox
steps. The CHPO process is based on the key step of propylene epox-
idation with CMHP, using mesoporous Ti-containing silica as the
catalyst. We thus investigated PO formation over LTS-1, using
CMHP in excess of propylene by comparing to other catalysts. As
shown in Table 7, the reaction was highly selective independent
of titanosilicates, giving PO as the sole product (>99%). In the case
of microporous titanosilicates, employing bulky CMHP as the oxi-
dant made only those Ti sites located on the external surface and/
or near pore entrance available to the reaction. Reasonably, CTS-1,
Ti-Beta, and Ti-MWW showed an extremely low PO yield (Table
7, Nos. 1–3). LTS-1, however, was highly active, showing a PO yield
even higher than that of Ti-MCM-41 (Table 7, Nos. 4 and 5). Based
on TOF, the activity of LTS-1 was about four times as high as that
of mesoporous Ti-MCM-41, implying that LTS-1 with mesoporosity
is a promising catalyst useful for the CHPO process.
[17] A. Corma, M.J. Diaz-Cabanas, J.L. Jorda, C. Martinez, M. Moliner, Nature 443
(2006) 842.
[18] T. Blasco, A. Corma, M.T. Navarro, J.P. Pariente, J. Catal. 156 (1995) 65.
[19] P. Wu, T. Tatsumi, T. Komatsu, T. Yashima, Chem. Mater. 14 (2002) 1657.
[20] J.C. Groen, T. Bach, U. Ziese, A.M. Paulaime-van Donk, K.P. de Jong, J.A. Moulijn,
J. Pérez-Ramírez, J. Am. Chem. Soc. 127 (2005) 10792.
[21] J.C. Groen, L.A.A. Peffer, J.A. Moulijn, J. Pérez-Ramírez, Micropor. Mesopor.
Mater. 69 (2004) 29.
[22] J.C. Groen, J.C. Jansen, J.A. Moulijn, J. Pérez-Ramírez, J. Phys. Chem. B 108
(2004) 13062.
[23] K. Egeblad, M. Kustova, S.K. Klitgaard, K. Zhu, C.H. Christensen, Micropor.
Mesopor. Mater. 101 (2007) 214.
[24] Y. Tao, H. Kanoh, K. Kaneko, J. Am. Chem. Soc. 125 (2003) 6044.
[25] A. Sakthivel, S. Huang, W. Chen, Z. Lan, K. Chen, T. Kim, R. Ryoo, A.S.T. Chiang, S.
Liu, Chem. Mater. 16 (2004) 3168.
[26] M. Choi, H.S. Cho, R. Srivastava, C. Venkatesan, D. Choi, R. Ryoo, Nat. Mater. 5
(2006) 718.
ˇ
[27] W.J. Roth, J. Cejka, Catal. Sci. Technol. 1 (2011) 43.
[28] A. Corma, V. Fornes, S.B. Pergher, T.L.M. Maesen, J.G. Buglass, Nature 396
(1998) 353.
4. Conclusions
[29] P. Wu, D. Nuntasri, J. Ruan, Y. Liu, M. He, W. Fan, O. Terasaki, T. Tatsumi, J. Phys.
Chem. B 108 (2004) 19126.
[30] L. Wang, Y. Wang, Y. Liu, L. Chen, S. Cheng, G. Gao, M. He, P. Wu, Micropor.
Mesopor. Mater. 113 (2008) 435.
[31] P. Wu, J. Ruan, L. Wang, L. Wu, Y. Wang, Y. Liu, W. Fan, M. He, O. Terasaki, T.
Tatsumi, J. Am. Chem. Soc. 130 (2008) 8178.
[32] M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki, R. Ryoo, Nature 461 (2009) 246.
[33] K. Na, C. Jo, J. Kim, W.S. Ahn, R. Ryoo, ACS Catal. 1 (2011) 901.
[34] P. Wu, T. Tatsumi, T. Komatsu, T. Yashima, J. Phys. Chem. B 105 (2001) 2897.
[35] G. Ricchiardi, A. Damin, S. Bordiga, C. Lamberti, G. Spanò, F. Rivetti, A. Zecchina,
J. Am. Chem. Soc. 123 (2001) 11409.
[36] F. Boccuzzi, S. Coluccia, G. Ghiotti, C. Morterra, A. Zecchina, J. Phys. Chem. 82
(1978) 1298.
[37] E. Astorino, J.B. Peri, R.J. Willey, G. Busca, J. Catal. 157 (1995) 482.
[38] Y. Wang, Y. Liu, X. Li, H. Wu, M. He, P. Wu, J. Catal. 266 (2009) 258.
[39] L. Wang, Y. Liu, W. Xie, P. Wu, J. Phys. Chem. C 112 (2008) 6132.
[40] J. Bu, H.K. Rhee, Catal. Lett. 66 (2000) 245.
Lamellar TS-1 with a multilayer structure has been synthesized
successfully with the aid of a bifunctional surfactant, C22H45AN+
(CH3)2AC6H12AN+(CH3)2AC6H13(OH)2. The new material preserves
the highly crystalline nature of conventional TS-1, and at the same
time, it features a large number of Ti active sites inside the mesop-
ores after calcination. In the epoxidation of cyclic alkenes with TBHP
or aqueous H2O2, LTS-1 is more active than microporous TS-1,
Ti-Beta, and Ti-MWW, and it is comparably active to mesoporous
Ti-MCM-41. The unique catalytic behavior of LTS-1 in bulky reac-
tions is mainly attributed to its mesoporosity and ultrathin crystal
morphology, which improves the accessibility of active Ti to the
substrate molecules. The relatively high hydrophilicity of LTS-1
makes it less active for the reactions using H2O2 oxidant. However,
in terms of catalytic activity, reuse, and stability of Ti species against
leaching, LTS-1 is characteristic of crystalline titanosilicates rather
than amorphous mesoporous materials. Moreover, LTS-1 proves
to be a promising catalyst for the production of PO through the
epoxidation of propylene with CMHP.
[41] Y. Cheneviere, F. Chieux, V. Caps, A. Tuel, J. Catal. 269 (2010) 161.
[42] Z. Zhao, Y. Liu, H. Wu, X. Li, M. He, P. Wu, J. Porous Mater. 17 (2010) 399.
[43] W. Fan, P. Wu, T. Tatsumi, J. Catal. 256 (2008) 62.
[44] N. Igarashi, K. Hashimoto, T. Tatsumi, Micropor. Mesopor. Mater. 104 (2007)
269.
[45] P. Wu, H. Sugiyama, T. Tatsumi, Stud. Surf. Sci. Catal. 146 (2003) 613.
[46] M. Ishino, J. Yamamoto, Catalysts Catal. (Shokubai) 48 (2006) 511.